Merge pull request 'cuttingedge' (!58) from cuttingedge into master

Reviewed-on: #58

.github/ISSUE_TEMPLATE/new_connector.yml

@@ -0,0 +1,23 @@
+name: New Connector Request
+description: Request a new site to be added
+title: "[New Connector]: "
+labels: ["New Connector"]
+body:
+  - type: input
+    attributes:
+      label: Website-Link
+      placeholder: https://
+    validations:
+      required: true
+  - type: checkboxes
+    attributes:
+      label: Is the Website free to access?
+      description: We can't support pay-to-use sites.
+      options:
+        - label: The Website is freely accessible.
+          required: true
+  - type: textarea
+    attributes:
+      label: Anything else?
+    validations:
+      required: false

README.md

@@ -21,11 +21,6 @@
   </p>
 </div>
 
-# Important for existing users:
-Tranga just had a complete rewrite. Old settings, tasks, etc. will not work.
-~~For the time being the docker-tag `latest` will be the old, discontinued branch.~~ `cuttingedge` is the active branch and
-will soon be moved to the `latest` branch (This is now the case). There is no migration-tool. Make a backup of old files.
-
 <!-- TABLE OF CONTENTS -->
 <details>
   <summary>Table of Contents</summary>
@@ -57,10 +52,12 @@ will soon be moved to the `latest` branch (This is now the case). There is no mi
 
 Tranga can download Chapters and Metadata from "Scanlation" sites such as 
 
-- [MangaDex.org](https://mangadex.org/)
-- [Manganato.com](https://manganato.com/)
-- [Mangasee](https://mangasee123.com/)
-- [MangaKatana](https://mangakatana.com)
+- [MangaDex.org](https://mangadex.org/) (Multilingual)
+- [Manganato.com](https://manganato.com/) (en)
+- [Mangasee.com](https://mangasee123.com/) (en)
+- [MangaKatana.com](https://mangakatana.com) (en)
+- [Mangaworld.bz](https://www.mangaworld.bz/) (it)
+- [Bato.to](https://bato.to/v3x) (en)
 - ❓ Open an [issue](https://github.com/C9Glax/tranga/issues)
 
 and trigger an scan with [Komga](https://komga.org/) and [Kavita](https://www.kavitareader.com/).  

Tranga/Jobs/JobBoss.cs

@@ -162,14 +162,21 @@ public class JobBoss : GlobalBase
                 new JobJsonConverter(this, new MangaConnectorJsonConverter(this, connectors)))!;
             this.jobs.Add(job);
         }
-        
+
         //Connect jobs to parent-jobs and add Publications to cache
         foreach (Job job in this.jobs)
         {
             this.jobs.FirstOrDefault(jjob => jjob.id == job.parentJobId)?.AddSubJob(job);
-            if(job is DownloadNewChapters dncJob)
+            if (job is DownloadNewChapters dncJob)
                 cachedPublications.Add(dncJob.manga);
         }
+
+        HashSet<string> coverFileNames = cachedPublications.Select(manga => manga.coverFileNameInCache!).ToHashSet();
+        foreach (string fileName in Directory.GetFiles(settings.coverImageCache))
+        {
+            if(!coverFileNames.Any(existingManga => fileName.Contains(existingManga)))
+                File.Delete(fileName);
+        }
     }
 
     private void UpdateJobFile(Job job)

Tranga/MangaConnectors/Bato.cs

@@ -0,0 +1,209 @@
+using System.Net;
+using System.Text.RegularExpressions;
+using HtmlAgilityPack;
+using Tranga.Jobs;
+
+namespace Tranga.MangaConnectors;
+
+public class Bato : MangaConnector
+{
+	public Bato(GlobalBase clone) : base(clone, "Bato")
+	{
+		this.downloadClient = new HttpDownloadClient(clone, new Dictionary<byte, int>()
+		{
+			{1, 60}
+		});
+	}
+
+	public override Manga[] GetManga(string publicationTitle = "")
+	{
+		Log($"Searching Publications. Term=\"{publicationTitle}\"");
+		string sanitizedTitle = string.Join(' ', Regex.Matches(publicationTitle, "[A-z]*").Where(m => m.Value.Length > 0)).ToLower();
+		string requestUrl = $"https://bato.to/v3x-search?word={sanitizedTitle}&lang=en";
+		DownloadClient.RequestResult requestResult =
+			downloadClient.MakeRequest(requestUrl, 1);
+		if ((int)requestResult.statusCode < 200 || (int)requestResult.statusCode >= 300)
+			return Array.Empty<Manga>();
+
+		if (requestResult.htmlDocument is null)
+		{
+			Log($"Failed to retrieve site");
+			return Array.Empty<Manga>();
+		}
+			
+		Manga[] publications = ParsePublicationsFromHtml(requestResult.htmlDocument);
+		Log($"Retrieved {publications.Length} publications. Term=\"{publicationTitle}\"");
+		return publications;
+	}
+
+	public override Manga? GetMangaFromUrl(string url)
+	{
+		DownloadClient.RequestResult requestResult =
+			downloadClient.MakeRequest(url, 1);
+		if ((int)requestResult.statusCode < 200 || (int)requestResult.statusCode >= 300)
+			return null;
+		if (requestResult.htmlDocument is null)
+		{
+			Log($"Failed to retrieve site");
+			return null;
+		}
+		return ParseSinglePublicationFromHtml(requestResult.htmlDocument, url.Split('/')[^1]);
+	}
+
+	private Manga[] ParsePublicationsFromHtml(HtmlDocument document)
+	{
+		HtmlNode mangaList = document.DocumentNode.SelectSingleNode("//div[@data-hk='0-0-2']");
+		if (!mangaList.ChildNodes.Any(node => node.Name == "div"))
+			return Array.Empty<Manga>();
+
+		List<string> urls = mangaList.ChildNodes
+			.Select(node => $"https://bato.to{node.Descendants("div").First().FirstChild.GetAttributeValue("href", "")}").ToList();
+		
+		HashSet<Manga> ret = new();
+		foreach (string url in urls)
+		{
+			Manga? manga = GetMangaFromUrl(url);
+			if (manga is not null)
+				ret.Add((Manga)manga);
+		}
+
+		return ret.ToArray();
+	}
+
+	private Manga ParseSinglePublicationFromHtml(HtmlDocument document, string publicationId)
+	{
+		HtmlNode infoNode = document.DocumentNode.SelectSingleNode("/html/body/div/main/div[1]/div[2]");
+
+		string sortName = infoNode.Descendants("h3").First().InnerText;
+		string description = document.DocumentNode
+			.SelectSingleNode("//div[contains(concat(' ',normalize-space(@class),' '),'prose')]").InnerText;
+
+		string[] altTitlesList = infoNode.ChildNodes[1].ChildNodes[2].InnerText.Split('/');
+		int i = 0;
+		Dictionary<string, string> altTitles = altTitlesList.ToDictionary(s => i++.ToString(), s => s);
+
+		string posterUrl = document.DocumentNode.SelectNodes("//img")
+			.First(child => child.GetAttributeValue("data-hk", "") == "0-1-0").GetAttributeValue("src", "").Replace("&amp;", "&");
+		string coverFileNameInCache = SaveCoverImageToCache(posterUrl, 1);
+
+		List<HtmlNode> genreNodes = document.DocumentNode.SelectSingleNode("//b[text()='Genres:']/..").SelectNodes("span").ToList();
+		string[] tags = genreNodes.Select(node => node.FirstChild.InnerText).ToArray();
+
+		List<HtmlNode> authorsNodes = infoNode.ChildNodes[1].ChildNodes[3].Descendants("a").ToList();
+		List<string> authors = authorsNodes.Select(node => node.InnerText.Replace("amp;", "")).ToList();
+
+		HtmlNode? originalLanguageNode = document.DocumentNode.SelectSingleNode("//span[text()='Tr From']/..");
+		string originalLanguage = originalLanguageNode is not null ? originalLanguageNode.LastChild.InnerText : "";
+
+		if (!int.TryParse(
+			    document.DocumentNode.SelectSingleNode("//span[text()='Original Publication:']/..").LastChild.InnerText.Split('-')[0],
+			    out int year))
+			year = DateTime.Now.Year;
+
+		string status = document.DocumentNode.SelectSingleNode("//span[text()='Original Publication:']/..")
+			.ChildNodes[2].InnerText;
+
+		Manga manga = new (sortName, authors, description, altTitles, tags, posterUrl, coverFileNameInCache, new Dictionary<string, string>(),
+			year, originalLanguage, status, publicationId);
+		cachedPublications.Add(manga);
+		return manga;
+	}
+
+	public override Chapter[] GetChapters(Manga manga, string language="en")
+	{
+		Log($"Getting chapters {manga}");
+		string requestUrl = $"https://bato.to/title/{manga.publicationId}";
+		// Leaving this in for verification if the page exists
+		DownloadClient.RequestResult requestResult =
+			downloadClient.MakeRequest(requestUrl, 1);
+		if ((int)requestResult.statusCode < 200 || (int)requestResult.statusCode >= 300)
+			return Array.Empty<Chapter>();
+
+		//Return Chapters ordered by Chapter-Number
+		List<Chapter> chapters = ParseChaptersFromHtml(manga, requestUrl);
+		Log($"Got {chapters.Count} chapters. {manga}");
+		return chapters.OrderBy(chapter => Convert.ToSingle(chapter.chapterNumber, numberFormatDecimalPoint)).ToArray();
+	}
+
+	private List<Chapter> ParseChaptersFromHtml(Manga manga, string mangaUrl)
+	{
+		// Using HtmlWeb will include the chapters since they are loaded with js 
+		HtmlWeb web = new();
+		HtmlDocument document = web.Load(mangaUrl);
+
+		List<Chapter> ret = new();
+
+		HtmlNode chapterList =
+			document.DocumentNode.SelectSingleNode("/html/body/div/main/div[3]/astro-island/div/div[2]/div/div/astro-slot");
+
+		Regex chapterNumberRex = new(@"Chapter ([0-9\.]+)");
+
+		foreach (HtmlNode chapterInfo in chapterList.SelectNodes("div"))
+		{
+			HtmlNode infoNode = chapterInfo.FirstChild.FirstChild;
+			string fullString = infoNode.InnerText;
+
+			string? volumeNumber = null;
+			string chapterNumber = chapterNumberRex.Match(fullString).Groups[1].Value;
+			string chapterName = chapterNumber;
+			string url = $"https://bato.to{infoNode.GetAttributeValue("href", "")}?load=2";
+			ret.Add(new Chapter(manga, chapterName, volumeNumber, chapterNumber, url));
+		}
+		
+		return ret;
+	}
+
+	public override HttpStatusCode DownloadChapter(Chapter chapter, ProgressToken? progressToken = null)
+	{
+		if (progressToken?.cancellationRequested ?? false)
+		{
+			progressToken.Cancel();
+			return HttpStatusCode.RequestTimeout;
+		}
+
+		Manga chapterParentManga = chapter.parentManga;
+		Log($"Retrieving chapter-info {chapter} {chapterParentManga}");
+		string requestUrl = chapter.url;
+		// Leaving this in to check if the page exists
+		DownloadClient.RequestResult requestResult =
+			downloadClient.MakeRequest(requestUrl, 1);
+		if ((int)requestResult.statusCode < 200 || (int)requestResult.statusCode >= 300)
+		{
+			progressToken?.Cancel();
+			return requestResult.statusCode;
+		}
+
+		string[] imageUrls = ParseImageUrlsFromHtml(requestUrl);
+
+		string comicInfoPath = Path.GetTempFileName();
+		File.WriteAllText(comicInfoPath, chapter.GetComicInfoXmlString());
+
+		return DownloadChapterImages(imageUrls, chapter.GetArchiveFilePath(settings.downloadLocation), 1, comicInfoPath, "https://mangakatana.com/", progressToken:progressToken);
+	}
+
+	private string[] ParseImageUrlsFromHtml(string mangaUrl)
+	{
+		DownloadClient.RequestResult requestResult =
+			downloadClient.MakeRequest(mangaUrl, 1);
+		if ((int)requestResult.statusCode < 200 || (int)requestResult.statusCode >= 300)
+		{
+			return Array.Empty<string>();
+		}
+		if (requestResult.htmlDocument is null)
+		{
+			Log($"Failed to retrieve site");
+			return Array.Empty<string>();
+		}
+
+		HtmlDocument document = requestResult.htmlDocument;
+
+		HtmlNode images = document.DocumentNode.SelectNodes("//astro-island").First(node =>
+			node.GetAttributeValue("component-url", "").Contains("/_astro/ImageList."));
+
+		string weirdString = images.OuterHtml;
+		string weirdString2 = Regex.Match(weirdString, @"props=\""(.*)}\""").Groups[1].Value;
+		string[] urls = Regex.Matches(weirdString2, @"https:\/\/[A-z\-0-9\.\?\&\;\=\/]*").Select(m => m.Value.Replace("\\&quot;]", "").Replace("amp;", "")).ToArray();
+		
+		return urls;
+	}
+}

Tranga/MangaConnectors/MangaConnector.cs

@@ -213,8 +213,7 @@ public abstract class MangaConnector : GlobalBase
         //Download all Images to temporary Folder
         foreach (string imageUrl in imageUrls)
         {
-            string[] split = imageUrl.Split('.');
-            string extension = split[^1];
+            string extension = imageUrl.Split('.')[^1].Split('?')[0];
             Log($"Downloading image {chapter + 1:000}/{imageUrls.Length:000}"); //TODO
             HttpStatusCode status = DownloadImage(imageUrl, Path.Join(tempFolder, $"{chapter++}.{extension}"), requestType, referrer);
             Log($"{saveArchiveFilePath} {chapter + 1:000}/{imageUrls.Length:000} {status}");
@@ -247,7 +246,8 @@ public abstract class MangaConnector : GlobalBase
     
     protected string SaveCoverImageToCache(string url, byte requestType)
     {
-        string filename = url.Split('/')[^1].Split('?')[0];
+        string filetype = url.Split('/')[^1].Split('?')[0].Split('.')[^1];
+        string filename = $"{DateTime.Now.Ticks.ToString()}.{filetype}";
         string saveImagePath = Path.Join(settings.coverImageCache, filename);
 
         if (File.Exists(saveImagePath))

Tranga/MangaConnectors/MangaConnectorJsonConverter.cs

@@ -34,6 +34,8 @@ public class MangaConnectorJsonConverter : JsonConverter
                 return this._connectors.First(c => c is Mangasee);
             case "Mangaworld":
                 return this._connectors.First(c => c is Mangaworld);
+            case "Bato":
+                return this._connectors.First(c => c is Bato);
         }
 
         throw new Exception();

Tranga/Tranga.cs

@@ -22,7 +22,8 @@ public partial class Tranga : GlobalBase
             new Mangasee(this),
             new MangaDex(this),
             new MangaKatana(this),
-            new Mangaworld(this)
+            new Mangaworld(this),
+            new Bato(this)
         };
         jobBoss = new(this, this._connectors);
         StartJobBoss();